Re: using DMA-API on ARM

[Arend van Spriel <arend@broadcom.com>]

On 12/05/14 15:20, Hante Meuleman wrote:
> Ok, I'll add the necessary debug to get all the information out,
> but it will take some time to get it done, so I won't have anything
> before Monday.
>
> -----Original Message-----
> From: Russell King - ARM Linux [mailto:linux@arm.linux.org.uk]
> Sent: vrijdag 5 december 2014 14:24
> To: Hante Meuleman
> Cc: Will Deacon; Arend Van Spriel; Marek Szyprowski; linux-arm-kernel@lists.infradead.org; David Miller; linux-kernel@vger.kernel.org; brcm80211-dev-list; linux-wireless
> Subject: Re: using DMA-API on ARM
>
> Please wrap your message - replying to a message which looks like this in
> my editor is far from easy, and gives me much more work to /manually/
> reformat it before I can reply to it:

That's what happens with corporate IT forcing to use Outlook. We can 
workaround that using Thunderbird on Citrix. I will enlighten Hante 
about that option :-)

Regards,
Arend

> On Fri, Dec 05, 2014 at 12:56:45PM +0000, Hante Meuleman wrote:
>> The problem is with data coming from device, so DMA from device to host. The $
>>
>> However: this indicates that dma_alloc_coherent on an ARM target may result i$
>>
>> Regards,
>> Hante
>
> Thanks.
>
> On Fri, Dec 05, 2014 at 12:56:45PM +0000, Hante Meuleman wrote:
>> However: this indicates that dma_alloc_coherent on an ARM target may
>> result in a memory buffer which can be cached which conflicts with
>> the API of this function.
>
> If the memory has an alias which is cacheable, it is possible for cache
> lines to get allocated via that alias, even if the alias has no explicit
> accesses to it.
>
> This is something which I've been going on for quite literally /years/ -
> mismatched cache attributes can cause unpredictable behaviour.  I've had
> a lot of push back from people who are of the opinion that "if it works
> for me, then there isn't a problem" and I eventually gave up fighting
> the battle, especially as the ARM architecture people weakened my
> reasoning behind it by publishing a relaxation of the "no differing
> attributes" issue.  This was particularly true of those who wanted to
> use ioremap() on system memory - and cases such as
> dma_init_coherent_memory().
>
> So, I never fixed this problem in the original DMA allocator code; I
> basically gave up with it.  It's a latent bug which did need to be fixed,
> and is still present today in the non-CMA case.
>
> The symptoms which you are reporting sound very much like this kind of
> problem - the virtual address for the memory returned by
> dma_alloc_coherent() will not be cacheable memory - it will have been
> remapped using map_vm_area().  However, there could very well be a fully
> cacheable lowmem mapping of that memory, which if a read (speculative or
> otherwise) will bring a cache line in, and because the caches are VIPT
> or PIPT, that cache line can be hit via the non-cacheable mapping too.
>
> What I /really/ need is more evidence of this to tell those disbelievers
> where to stick their flawed arguments. :)
>